Abstract The International Mouse Phenotyping Consortium (IMPC) has developed an invaluable repository of gene knockout (KO) mice and has performed a whole-animal assessment of each KO line. Approximately 1/3 of the KO lines are embryonic lethals that were characterized as heterozygous (HET) KO lines. Previously we performed high throughput CT and histomorphometric analysis on 220 unselected viable homozygous (HOM) KO lines. This study found that ~12.5% have an unequivocal variance in total bone mass (trabecular bone volume and/or bone size), most of which were not detected by the IMPC screening. The results and interpretation of the KOs examined in the screen are available on a web portal (bonebase.org). Based on this experience, the phenotyping data on the IMPC webportal and the literature, we estimate that the HET population of KOs will be an equally rich source of genes that affect bone/body variance and skeletal health in the later adult years. This proposal will apply our skeletal phenotyping workflow to selected HET KO lines that are likely to have a significant bone/body mass phenotype. Both CT and total body composition as determined by TD-NMR will be used as the screening modality. HET KO lines with significant variance in bone architecture will be processed for histomorphometry. Based on those results, certain lines will undergo studies of osteogenesis (Raman microscopy for the mineral/matrix composition, primary cell culture for osteoblast) and/or osteoclastogenesis (collagen crosslinks and primary cell culture for osteoclast). In HET KO lines with variance in body composition, histomorphometry of the skeletal muscle will be performed. The intent is to discriminate the impact of the HET KO as acting directly on the osteoblast or osteoclast, or whether the observed phenotype is a secondary effect of the HET KO on other tissues that can influence bone. In addition, HOM KO lethal lines that survive to 18.5 days of gestation will undergo a gross and histological skeletal examination to determine the impact of the gene on skeletal development. To complement these biological findings, a bioinformatics component is being added to map the KO genes that are affecting osteoblast or osteoclast directly to known or postulated molecular pathways. Does the gene participate in the function of the network or is it a product of a network? The final goal of the project is to place the biological and bioinformatic data into a classification structure that reveals differences and similarities between the various KO lines. We will engage internal and external experts in bone biology to review the data assembled for a specific HET KO line to critique and amplify our interpretation and ever-evolving classification schema. Recognizing that frequency and complexity of the genes affecting bone is a big-data challenge (estimated to be ~3,500 genes), our intent is to lay the ground work for how this information will be gathered, presented, interpreted, queried and eventually applied toward the goal of personalized skeletal human health.